Training compact deep learning models for video classification using circulant matrices

Abstract

In real world scenarios, model accuracy is hardly the only factor to consider. Large models consume more memory and are computationally more intensive, which makes them difficult to train and to deploy, especially on mobile devices. In this paper, we build on recent results at the crossroads of Linear Algebra and Deep Learning which demonstrate how imposing a structure on large weight matrices can be used to reduce the size of the model. We propose very compact models for video classification based on state-of-the-art network architectures such as Deep Bag-of-Frames, NetVLAD and NetFisherVectors. We then conduct thorough experiments using the large YouTube-8M video classification dataset. As we will show, the circulant DBoF embedding achieves an excellent trade-off between size and accuracy.

The 2nd Workshop on YouTube-8M Large-Scale Video Understanding, ECCV 2018

References in corpus (13)

• Distilling the Knowledge in a Neural Network
• BinaryConnect: Training Deep Neural Networks with binary weights during propagations
• Deep Learning with Limited Numerical Precision
• YouTube-8M: A Large-Scale Video Classification Benchmark
• Predicting Parameters in Deep Learning
• Compressing Neural Networks with the Hashing Trick
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• Exploiting Feature and Class Relationships in Video Categorization with Regularized Deep Neural Networks
• Memory Bounded Deep Convolutional Networks
• Temporal Modeling Approaches for Large-scale Youtube-8M Video Understanding
• The Monkeytyping Solution to the YouTube-8M Video Understanding Challenge
• ACDC: A Structured Efficient Linear Layer
• Deep Learning Methods for Efficient Large Scale Video Labeling